Methods- Marginally Off Topic

  • Thread starter Thread starter SLuiCe
  • Start date Start date
This is like cooking

Y'know, I worked in a bunch of restaurant kitchens early on, and at one point I got a "Joe" job doing the short order stuff at a fancy club where there was a real chef and su checfs etc... (yes I'm going somewhere with this). One of the things that struck me was that fancy ass coooking isn't some carefull blend of 7 herbs and spices yada yada. Sure, the saucier made some special stuff, but by and large it was about starting off with good food and making sure the goodness carried through to the final product.

This conversation sort of reminded me of that. We should all make sure to keep our food stuffs fresh.
 
"WTF is 32 float???"

Its a 32 bit floating point number.

A simple and entirely fictitious example is a 5 digit number where the first 4 digits are the quantity and the last one is where to put the decimal point.

therefore 33330 would be 3333
and 33333 would be 3.333
33339 would be 0.000003333

The last case is like a slider set really low, you can imagine that a float could still describe the wave form, whereas 16 bit audio, which is a range from -32767 to 32768 would round the wave amplitude to a 0.

It quantifies large and small umbers with equal precision relative to the size of the number. Of course it does this in bits and not numbers.
 
Doug H said:
"WTF is 32 float???"

Its a 32 bit floating point number.
Click to expand...

A search can find plenty of info on floating point formats, but the essence of it is that the data is not truncated during processing the way it is in 16 bit, so you're not losing information in the waveform.

Even if you're only capable (as I am) of recording in 16 bit, it's still advisable to do all your mixing and processing in a 32 bit floating point format, if your software supports it. This (as the other chris noted) *will* have a discernable affect on the overall sound quality. Then, as the very last step, you dither back down to 16 bit.

This is not by any means a "Cool Edit specific" concept. It just seems that CE users are often more hip to what's going on, internally, in their software.

Chris
 
Yes, we hijacked this thread. But it's been a good discussion, and anyway, the thread title's "Marginally Off Topic". So we're sort of parallel to the spirit of the thread, right? :D

Chrisharris - I never do submixes. I don't know how. And as for that other thing you mentioned - the aux bus - (and all the other stuff you mentioned), I think it's time for me to study and learn. You're one of those people who are light years ahead of me.

What's a mixer?
 
Doug H said:
"WTF is 32 float???"

Its a 32 bit floating point number.

A simple and entirely fictitious example is a 5 digit number where the first 4 digits are the quantity and the last one is where to put the decimal point.

therefore 33330 would be 3333
and 33333 would be 3.333
33339 would be 0.000003333

The last case is like a slider set really low, you can imagine that a float could still describe the wave form, whereas 16 bit audio, which is a range from -32767 to 32768 would round the wave amplitude to a 0.

It quantifies large and small umbers with equal precision relative to the size of the number. Of course it does this in bits and not numbers.
Click to expand...

see all that damn math..THAT is why i will never be a good "mixer" i mean thats waaaaaaay too much math for rock music...:D
 
Jamal Bucket said:
see all that damn math..THAT is why i will never be a good "mixer" i mean thats waaaaaaay too much math for rock music...:D
Click to expand...

That math is why so many good engineers still prefer to mix on a good analog mixer.

That and there's something about actually turning a knob or sliding a fader that just feels good.
 
M.Brane said:
That and there's something about actually turning a knob or sliding a fader that just feels good.
Click to expand...
Ya' know.....that's the way I've been doing it for so long.
I have no fear of the digital mixers......I know that I could understand and use them just fine......but when I'm doing a mix, I just like to be able to reach up and grab a big ol' handful of knobs.
 
Thanks fellas,...................


I still don't understand though....
My Soundcard is 24 bit capable...
CEP only has 16 or 32 for the settings for tracking....besides stereo or mono that is....

the mix is set at 32....

Is this the best I can do?

Joe
 
haha...I can assure you I had no idea the mathmeticians would come out for this! This might possibly be causing some of that overwhelming feeling for some folks that I mentioned in the original post. I know I personally stopped reading like 2 pages ago. :)
 
Last edited:
New thread, new thread!

We should gather all the useful tips and paste them into one big thread (one you can't post in)...
 
Good stuff for everyone here! One thing though..its meant to be like a guide for newbies..and while most are giving great suggestions, the details arent there. For instance...
Sluice-- The point you made about making eq cuts for instruments to fit in nicer etc is great advice..and makes sense, but how does a newbie exactly go about this??

EQ--For the un-knowing newbie using soudforge or something..they'll record a track..say guitar, go the eq..find a preset called 'Guitar' and go..yeah that sounds better. If only it is that easy!

I think a newbie guide to making correct eq adjustments would be a great idea..though i'm not fit to babble about it.
 
joro said:
Chris.....

WTF is 32 float???:confused: :confused:

I record all my tracks at 16bit.....am I stupid?? or what??



...I still don't understand though....
My Soundcard is 24 bit capable...
CEP only has 16 or 32 for the settings for tracking....besides stereo or mono that is....
the mix is set at 32....
Is this the best I can do?
Click to expand...

No, you're not dumb...at least, not understanding the 16/24/32bit thing doesn't make you dumb... :D - I don't really understand it either.

However, stuff sounds better overall if you stick it all at 32bit. There's no such thing as 32 bit, really...I mean, there's no soundcard that supports "32 bit," but the way CEP (and other proggies) work is in "32 bit float," and Doug H explained it...I still don't understand it, but he explained it, lol.

Actually, I understand it some...but the bottom line is that if you work in 32 bit, no matter WHAT kind of soundcard you have, the end result will be better. I don't know about other programs, but working in CEP in 16 bit has the effect of attenuating the highs (and other stuff, I'm sure) b/c of all the unnecessary dithering going on.
 
BrettMckinney said:
Good stuff for everyone here! One thing though..its meant to be like a guide for newbies..and while most are giving great suggestions, the details arent there. For instance...
Sluice-- The point you made about making eq cuts for instruments to fit in nicer etc is great advice..and makes sense, but how does a newbie exactly go about this??

EQ--For the un-knowing newbie using soudforge or something..they'll record a track..say guitar, go the eq..find a preset called 'Guitar' and go..yeah that sounds better. If only it is that easy!

I think a newbie guide to making correct eq adjustments would be a great idea..though i'm not fit to babble about it.
Click to expand...

I just stopped home for a lunch break. I'll try to do something with EQ in the next couple days. I'm pressed fore time as it is to try and get the naked recording done in time for WATYF's deadline. Probably ain't gonna make it.

This 32 bit float stuff is all Greek to me. Maybe it would help me improve my own sound. I don't know. But I really was hoping this thread would be more of a benefit to newbies.

So if any beginners would like to ask some more beginner related questions, please feel free. Maybe we can take the thread back over.... :D
 
chrisharris said:
Oh, and it's a lot worse in CEP if you work in 16 bit, b/c then each and every process involves 2 additional conversions (from 16 bit to 32 bit, then back to 16 bit again)...it does this invisibly, and I killed a bunch of fidelity for a long time doing this. It's funny, but the first song I posted in the clinic after I started working in 32 float, most of the responses were like..."YOU GOT NEW GEAR!!"...lol.
Click to expand...

OK - I don't work in CEP so my questions may not make any sense or even apply. I don't understand why you have additional conversions. Doesn't that depend on what bit resolution you recorded the data in the first rate at. I mean if you record it a 16bit, mix it at 16bit and burn it to a CD (obviously at 16bit) then technically there is no dither, correct? Of course, are we talking about something different as I read more? Are you strictly talking about mixing mathmatics? I record at 24bit, mix at 24bit and dither to 16bit in PT. I don't have any other conversions.

I guess I don't get the multiple conversions thing. :)
 
Here's the best I can do to simplify. Anyone who wants a more detailed layout (and a more reputable source than me) just needs to do a web search. This info is basic stuff that's all over the web.

An example of the idea behind "floating point' would be this: say you have a wav file that you amplify past the point of clipping. With a 16 bit file, once you go that far, those bits of data that exceed 0 db are gone forever. You can't get them back. With 32 bit those bits are "preserved", so that whatever processing you do to a wav, the data is not truncated or clipped off the way it is in a fixed-point format like 16/44.1.

So (I'll say once again), even if you're only capable of recording in 16 bit and even if you're eventually planning to dither back down to 16 bit, you will be much better off converting to a floating point format while you do your processing. It's really not that complicated, and you don't need to get math involved at all.

Then there's the issue of the internal processing format of your software. I don't know doodly about Pro tools - I suspect their internal format is 24 bit. But it's good to have your files consistant with the internal mixing format of the software. Otherwise whenever you do anything, there are tiny little dithers happening.

It ain't rocket science if *I* can understand it, believe me....

Chris
 
SLuiCe said:
I just stopped home for a lunch break. I'll try to do something with EQ in the next couple days. I'm pressed fore time as it is to try and get the naked recording done in time for WATYF's deadline. Probably ain't gonna make it.

This 32 bit float stuff is all Greek to me. Maybe it would help me improve my own sound. I don't know. But I really was hoping this thread would be more of a benefit to newbies.

So if any beginners would like to ask some more beginner related questions, please feel free. Maybe we can take the thread back over.... :D
Click to expand...
Yeah... it just occured to me how little time that is....


that's OK... I'll just blame it on Chris. :p


Don't worry aboot the deadline... people are already submitting tunes... and we'll prolly have a steady stream of submission for a few weeks. I guess a deadline is kinda irrelevant.


WATYF
 
I would say Groucho has summed up the floating point theory quite well.

This concludes our weekly Calculus lesson.

EQ--For the un-knowing newbie using soudforge or something..they'll record a track..say guitar, go the eq..find a preset called 'Guitar' and go..yeah that sounds better. If only it is that easy!
Click to expand...

Brett- I totally I agree. And I know very well that there's a lot of people using those presets, and I am constantly trying to discourage that. The only explanation for that is laziness. You have to get into it if you are going to be your own engineer. I can't stand the mathmatics of it all either, and like anyone, I was overwhelmed in the beginnning, but I know it's the only way I am going to get to hear my music the way it is in my head. And I don't mean you have to understand how many ones and zeros there are in g major chord if you use Cool Edit Pro or Cakewalk version 2.0. Like I said, the point of my post was to help keep people from feeling overloaded.

I'm going to try to combine a little EQ "tutorial" with my naked mix submission. I may even take it a step further and also provide some reverb details. I've got some really neat ideas about how to do that and I just need some time to work it out. But basically I'm hoping to find the time to provide comparison samples between my raw tracks and mixed tracks, and give specifics about how I went about choosing my settings. After that, I'm leaving this board forever.
 
SLuiCe said:
I would say Groucho has summed up the floating point theory quite well.

This concludes our weekly Calculus lesson.





I'm going to try to combine a little EQ "tutorial" with my naked mix submission. I may even take it a step further and also provide some reverb details. I've got some really neat ideas about how to do that and I just need some time to work it out. But basically I'm hoping to find the time to provide comparison samples between my raw tracks and mixed tracks, and give specifics about how I went about choosing my settings. After that, I'm leaving this board forever.
Click to expand...


I'm looking very forward to this!
The tutorial, not you leaving! lol

And you are completely right about people being lazy and using the presets. I do that! lol

But I'm learning and have been reading a lot of articles on EQ, good article in the current SOS mag but that mag is expensive in the sates. One thing that I have struggled with is reading an article and having it make sense to me in terms of sound. That's where some trial and error can come in but you have to get your feet wet!
They don't call it engineering because it's simple, we all know you have to dedicate your life to this stuff. Beats the hell out TV!
 
You must log in or register to reply here.
Back
Top